To test a bipolar junction transistor (BJT) with a multimeter, set the dial to Diode Test mode and measure the forward voltage drop across the base-emitter and base-collector junctions. A good silicon NPN transistor will read between 0.500V and 0.800V in the forward direction and show an open circuit (OL) in reverse. Testing a MOSFET requires checking the internal body diode and using the meter's internal voltage to charge the gate capacitance. You cannot reliably test these components using the standard resistance or continuity settings.
Meter Setup and Safety Categories for Bench Testing
Before touching any probes to silicon, configure your digital multimeter (DMM) correctly. The resistance (Ω) mode applies a variable test current that makes junction readings ambiguous. The Diode Test mode sources a constant current (typically 1mA to 2mA) and measures the exact forward voltage drop, which is the only reliable way to evaluate semiconductor junctions.
- Dial Position: Diode Test (symbol: a diode with a plus sign
->|+or similar). - Lead Jacks: Black lead to
COM. Red lead toV/Ω/mA(never the high-current 10A jack). - Range: Auto-ranging is standard. If manual, set to the lowest DC voltage range that exceeds 2V.
- Open-Circuit Verification: Touch probes together. The meter should read 0.000V or emit a continuity beep. Separate them; it should read
OL(Over Limit).
If you are testing a transistor out-of-circuit on a bench, a standard CAT II rated meter is sufficient. However, if you are probing transistors in-circuit inside a mains-powered Switch-Mode Power Supply (SMPS), motor drive, or inverter, your meter and probes must be rated CAT III 1000V or CAT IV 600V per IEC 61010 standards. Furthermore, you must physically discharge the bulk DC bus capacitors with a high-wattage bleeder resistor before testing. A charged 400V DC bus will instantly destroy a CAT II meter and can cause catastrophic arc flashes. Always verify dead with a proven meter before probing in-circuit.
BJT Junction Testing: Probe Placement and Expected Values
A Bipolar Junction Transistor (BJT) functions electrically as two diodes sharing a common anode (NPN) or cathode (PNP). By treating the Base as the common terminal, we can map the health of the silicon. The following table assumes standard silicon transistors (like the 2N2222, 2N3904, or TIP31). Germanium transistors will show much lower forward drops (0.200V - 0.300V).
| Transistor Type | Junction Tested | Red Probe (+) | Black Probe (-) | Expected Good Reading | Expected Bad Reading |
|---|---|---|---|---|---|
| NPN | Base-Emitter (Fwd) | Base | Emitter | 0.550V - 0.750V | < 0.400V or OL |
| NPN | Base-Collector (Fwd) | Base | Collector | 0.550V - 0.750V | < 0.400V or OL |
| NPN | Emitter-Base (Rev) | Emitter | Base | OL (Open) | Any numeric value |
| NPN | Collector-Emitter | Collector | Emitter | OL (Open) | Any numeric value |
| PNP | Emitter-Base (Fwd) | Emitter | Base | 0.550V - 0.750V | < 0.400V or OL |
| PNP | Collector-Base (Rev) | Collector | Base | OL (Open) | Any numeric value |
According to Fluke's electronics testing guidelines, identifying an unknown BJT's pinout relies on finding the Base first. Probe pins randomly until you find one pin that yields a 0.6V drop to both of the other two pins when the red probe is on it. That pin is the Base, and the transistor is NPN. If the black probe must be on the common pin to get the 0.6V readings, it is a PNP.
Power MOSFET Testing: Body Diode and Gate Charge
MOSFETs (like the ubiquitous IRFZ44N or IRLB8721) do not have base-emitter diodes. Instead, they feature an intrinsic body diode between the Drain and Source, and a highly capacitive Gate. Testing a MOSFET requires a two-step process: verifying the body diode and actively turning the channel on and off using your meter's internal battery.
- Discharge the Gate: Touch the black probe to the Gate and the red probe to the Source. This bleeds off any residual gate charge, ensuring the MOSFET is fully OFF.
- Test the Body Diode (OFF state): Place the red probe on the Drain and the black probe on the Source. You should read
OL. Reverse the probes (Red on Source, Black on Drain). You should read the body diode forward voltage drop, typically between 0.400V and 0.600V. - Charge the Gate (Turn ON): Keep the black probe on the Source. Move the red probe from the Drain to the Gate. The meter's internal 3V to 9V battery will charge the gate capacitance, turning the N-channel MOSFET ON.
- Verify the Channel (ON state): Move the red probe back to the Drain (keep black on Source). The meter should now read a very low voltage drop, typically 0.000V to 0.050V, indicating the channel is conducting.
- Discharge to Turn OFF: Repeat Step 1. Re-test the Drain-Source path as in Step 2 to confirm it has returned to
OLin the forward direction.
If the MOSFET fails to turn on in Step 4, or fails to turn off after discharging the gate, the silicon is compromised. For deeper analysis of MOSFET gate threshold voltages, refer to the semiconductor chapters in All About Circuits, which detail the physics of junction capacitance and threshold limits.
Five Mistakes That Give Misleading Readings
Even with the correct dial setting, bench environment variables and circuit topology can trick you into throwing away good parts or installing bad ones.
1. Testing In-Circuit Without Isolating Parallel Paths
A transistor soldered to a PCB is surrounded by snubber networks, transformer windings, and bleeder resistors. If you measure 0.450V across a collector-emitter junction in-circuit, the transistor might be perfectly fine; you are simply reading the forward drop of a parallel flyback diode or a low-value shunt resistor. Always lift at least one leg of the transistor, or desolder it entirely, before passing judgment.
2. The Skin Resistance Error
If you hold the transistor body and the metal probe tips with your bare fingers simultaneously, your skin's resistance (typically 50kΩ to 200kΩ depending on moisture) creates a parallel path. On high-impedance reverse-bias junction tests, this can pull an OL reading down to a misleading 1.5V or 2.0V, making you think the junction is leaky. Use alligator clips or a PCB vise to hold the probes.
3. Relying on the hFE Socket for Pass/Fail
Many multimeters have a transistor hFE (DC current gain) socket. While useful for sorting matched pairs of small-signal transistors (like grouping 2N3904s by gain), it is useless for fault-finding. The socket applies a tiny base current and a low collector voltage (usually under 3V). A transistor with a severely degraded collector-base junction might still show a 'normal' hFE at 3V, but will avalanche and short out instantly when subjected to its rated 40V or 60V operating voltage in a real circuit. Stick to the diode test for health checks.
4. Using Continuity Mode Instead of Diode Mode
Continuity mode only checks if resistance is below a threshold (usually 15Ω to 50Ω) and beeps. It will not display the 0.650V forward drop of a silicon junction. If you use continuity mode, a good transistor and a dead short might both trigger the beep, or a good junction might register as an open circuit because its 650mV drop exceeds the meter's continuity threshold voltage.
5. Forgetting P-Channel MOSFET Logic
The MOSFET gate-charge trick described above applies to N-Channel enhancement MOSFETs. If you are testing a P-Channel MOSFET (like the IRF9540N), the polarities are reversed. You must use the red probe to discharge the gate to the source, and the black probe to charge the gate relative to the source to turn it on. Applying N-channel logic to a P-channel part will yield confusing OL readings and lead to false condemnations.
